A method for measuring distance based on dual paths of photoelectric encoder

　　The five-wheel instrument method is to install a fifth wheel on the measured vehicle that is flexible in rotation and reliably in contact with the ground with a certain pressure (when measuring the speed of a tractor, the front wheel of the tractor is often used as the fifth wheel directly). When the vehicle moves, the ground friction can cause the attached fifth wheel to rotate. Using a photoelectric switch or a Hall switch and corresponding devices, several pulses can be generated per rotation. These pulses can be used to detect the wheel speed and running distance. The traditional five-wheel instrument method is a typical single-path contact measurement method. The hardness, friction coefficient, contact pressure, etc. of the ground surface will affect the measurement results, mainly in two aspects: First, due to the soft ground, the wheel sinks, resulting in the effective diameter of the single wheel rotating due to friction and the actual diameter may be inconsistent. Second, the uneven ground causes the length of the path rolled by the wheel to be inconsistent with the straight-line distance. For example, the machine actually moves forward 1m, and the wheel just rolls over a 45° peak or pit, so the reflected distance is about 1.4m. If accurate measurement results are required, the influence of the first type of problem can be eliminated by calibrating the pulse and arc length equivalent for each specific measured terrain condition, but there is no way to completely eliminate the second type of problem. The method proposed in this paper can effectively weaken the influence of uneven ground on measurement results.

　　2. Measurement Methods and Application Algorithms

　　Using the idea of ​​converting linear motion into rotation using the five-wheel instrument, rotary encoders of the same model are installed on the wheels (five-wheel instrument) on both sides of the traveling device, and are made to rotate synchronously with the rotating mechanism (wheel) through special accessories. The two wheels of this device are spaced 1m apart to ensure that the positioning signal can be fully collected between them, so as to avoid the phenomenon of missing position signals during detection, which will cause avoidable errors in spacing measurement. The control panel on the operating device is used to start the collection of external signals, and the position signal collected by the single-chip system is used as the basis for external interruption to control the start and stop of the encoder pulse signal counting, and the result of each count is saved to the specified storage unit.

　　The so-called spacing measurement refers to the displacement of two adjacent position signals. Traditional five-wheel instruments are all single-channel measurements, and only use the running conditions of one channel to determine the results. If the conditions of this channel are very complicated, there will be a deviation between the reflected distance and the actual displacement. Considering that the uneven surface mainly includes two situations, peaks and pits, no matter which situation affects the distance shown by the wheel running, it will increase. In order to effectively overcome the disadvantage of single-channel measurement being too one-sided, it is changed to dual-channel simultaneous detection. Due to the differences in the surface conditions of each route, for example, if one route has a spike or pit within a certain distance and the other route is straight, then the detection result must be that the distance reflected by the second route is shorter than that of the first route, and the corresponding output pulse number is less than that of the first route, and at the same time it is greater than or equal to the number of pulses corresponding to the actual value, then the measurement result of the second route will be closer to the true value; if both routes encounter spikes or pits during travel, the measurement results will also differ due to the different travel distances. Although this situation will also produce a large error in the measurement of the two routes, we can still determine that the route with fewer output pulses is closer to the true value; if both routes are straight, then the measurement results of the dual route and the single route will not be much different, and the error will not be too large no matter which route is selected. The advantages of dual-channel measurement are more obvious in the first two cases.

　　The above discussion can lead to the following algorithm:

　　(1) Compare the two pulse numbers obtained between adjacent position signals. If there is a difference, save the smaller value; if there is no difference, select any value (called the optimal value selection) as the final conversion spacing data (as shown in Figure 1);

　　(2) Since each pulse corresponds to a certain angle of rotation of the encoder, the distance traveled by the corresponding transmission mechanism is obtained by using the relationship between the angle and the circumference of the wheel rim, that is, the measured spacing. The formula is: is the total number of pulses output when the encoder rotates one circle, is the number of pulses at the nith interruption, r is the outer radius of the transmission mechanism, and s is the spacing.

　　Since the starting points of the two encoders may be inconsistent (depending on the design structure of the encoder itself), when the rising edge of one pulse arrives, the rising edge of the other pulse does not arrive. When counting pulses, there will be an error of 1/N (N is the number of pulses output by the encoder in one circle) between the two. The larger the N value, the smaller the error. Therefore, a high-resolution encoder can be used to reduce this impact.

　　3. Test and result analysis

　　A double-channel and single-channel comparison test was conducted in the field, using 51 infrared signals arranged at equal longitudinal spacing (S=0.5m) as position signals. It is known that the diameter of the traveling wheel is 50cm, and the encoder used is  the E6B2-CWZ6C encoder produced by Omron , which has an output pulse number of 360p/r, that is, the forward displacement of the device is 1.57m for one rotation of the wheel.

　　The results are shown in Table 1:

　　(1) The following can be obtained from the test data:

　　The dotted line and solid line represent the single-channel and dual-channel measurement data waveforms, respectively. From this chart, we can intuitively see that dual-channel measurement alleviates some of the peaks that appear in single-channel measurement. Some of the measured values ​​are smaller than the actual values, which should be due to the slippage of the two wheels during driving. In this case, taking a small value will make the measurement result deviate further from the accurate value, but considering that the probability of this happening in 50 intervals is only 4%, it is still obvious that the method of dual-channel measurement and optimal value selection makes the result close to the true value.

　　(2) Through the following one-way analysis of variance Table 2

　　Because F=13.67>F0.01(1,98)=6.93, it is considered that the difference between the single-channel and dual-channel methods is extremely significant.

　　(3) Through the following variance analysis box Figure 3 (1 is single-channel, 2 is dual-channel), it can also be concluded that the dual-channel measurement result is closer to the true value than the single-channel measurement result.

　　IV. Conclusion

　　The innovation of this paper is that it changes the traditional five-wheel instrument single-channel measurement method which is greatly affected by the surface conditions. It adopts this method of dual-channel simultaneous measurement and optimal selection of the results to effectively reduce the error. A series of tests provide a strong basis for this demonstration. Since this method uses the position signal to trigger the external interruption start and stop counting, as long as the positioning signal of the measured object is detected, the non-contact real-time measurement of the distance between the measured objects can be achieved. Therefore, this method has a positive guiding role in realizing the non-contact detection of the distance between underground seed grains.